Maarten

It looks like your right Stefan. In my zeal to show that there is a bug in XFLR5, I extended the winglets by 1 meter. Well, the angle of attack of the winglets dropped nicely to zero about half way through. So the difference must be in the transition between the wing to the winglet, like you already said. I have tried different transitions. It gets better, especially when the dihedral turn is made before the twist is reduced. Also an offset of the winglet with 30% of the chord helps (just as in the...

Indeed, stability is the main issue by flying wings. For planks, mostly flying at a low Cl, the viscous drag is dominant. I mostly design and build flying wings with a sweep ( I could not resist to attach a picture. No its not flying upside down ;-) Sweep wings use 'normal' airfoils and fly at a higher Cl, where the induced drag ( and the spanload) is also important. I use the program FLZ vortex as well. As it uses the Eppler code to calculates the airfoils , it is less accurate on the Cm of the...

Indeed, stability is the main issue by flying wings. For planks, mostly flying at a low Cl, the viscous drag is dominant. I mostly design and build flying wings with a sweep ( I could not resist to attach a picture. No its not flying upside down ;-) Sweep wings use 'normal' airfoils and fly at a higher Cl, where the induced drag ( and the spanload) is also important. I use the program FLZ as well. As it uses the Eppler code to calculates the airfoils , it is less accurate on the Cm of the airfoils...

Indeed, stability is the main issue by flying wings. For planks, mostly flying at a low Cl, the viscous drag is dominant. I mostly design and build flying wings with a sweep ( I could not resist to attach a picture. No its not flying upside down ;-) Sweep wings use 'normal' airfoils and fly at a higher Cl, where the induced drag ( and the spanload) is also important. I use the program FLZ as well. As it uses the Eppler code to calculates the airfoils , it is less accurate on the Cm of the airfoils...

Thank you Stefan for the explanation. I understand that 'total angle' is only an internal parameter, so no problem there. The Cl curve of the both identical plane's difference however also a lot, see the attachment. And with it the load distribution over the span. Which one should be followed to design an elliptical spanload?

And the xfl file

Thank you for your quick response Stefan. And your right, i forgot to add a section at this test model between the wing and the winglet to separate the twist and airfoil between them. However, the problem stays the same. Even with an additional 45 deg section between the wing and winglet. The total angle of the winglet moves with the wing, what is not correct. See the attached file

Thank you for your quick response Stefan. And your right, i forgot to add a section at this test model between the wing and the winglet to separate the twist and airfoil between them. However, the problem stays the same. Even if a add an additional 45 deg section between the wing and winglet. The total angle of the winglet moves with the wing, what is not correct. See the attached file